Method of determining color composition of an image

Claim 1

A method of determining color composition of an image, the method comprising: calculating an intensity value and a saturation value for each pixel of the image, wherein the intensity value of each pixel is calculated by the formula I=(R+G+B)/3, where I represents intensity value, R, G and B respectively represent red, green, and blue color levels, and the saturation value of each pixel is calculated by the formula S=1−Min(R, G, B)/I, where S represents the saturation value and Min(R, G, B) selects the minimum color level among the R, G, and B color levels; comparing the calculated intensity and saturation values for each pixel with first and second predetermined threshold values, respectively; labeling pixels with calculated intensity values above the first predetermined threshold value and calculated saturation values above the second predetermined threshold value as color pixels; applying a mask to the image and counting the number of color pixels out of the pixels selected by the mask; and determining that the image is a color image if the number of color pixels selected by the mask is greater than or equal to a predetermined density value.

Claim 2

A method of determining color composition of an image, the method comprising: calculating an intensity value and a saturation value for each pixel of the image, wherein the intensity value of each pixel is calculated by the formula I=(R+G+B)/3, where I represents intensity value, R, G and B respectively represent red, green, and blue color levels, and the saturation value of each pixel is calculated by the formula S=1−Min(R, G, B)/I, where S represents the saturation value and Min(R, G, B) selects the minimum color level among the R, G, and B color levels; comparing the calculated intensity and saturation values for each pixel with first and second predetermined threshold values, respectively; labeling pixels with calculated intensity values above the first predetermined threshold value and calculated saturation values above the second predetermined threshold value as color pixels; applying a mask to the image and counting the number of color pixels out of the pixels selected by the mask; and determining that the image is a color image if the number of color pixels selected by the mask is greater than or equal to a predetermined density value.

Claim 5

A method of determining color composition of an image, the method comprising: calculating an intensity value and a saturation value for each pixel of the image; comparing the calculated intensity and saturation values for each pixel with first and second predetermined threshold values, respectively; labeling pixels with calculated intensity values above the first predetermined threshold value and calculated saturation values above the second predetermined threshold value as color pixels; applying a mask to the image and counting the number of color pixels out of the pixels selected by the mask; determining that the image is a color image if the number of color pixels selected by the mask is greater than or equal to a predetermined density value; and after determining that the image is not a color image, the method further comprises: calculating a first histogram of the intensity values of all of the pixels in the image, the first histogram being divided into a first predetermined number of intensity ranges; choosing an intensity range in the first histogram containing the greatest number of pixels; setting a median value of the chosen intensity range as a background value for the image; updating the intensity values of the pixels in the image by performing a dilation function if the background value is greater than a third predetermined threshold value or performing an erosion function if the background value is less than or equal to the third predetermined threshold value; calculating a second histogram of the updated intensity values of all of the pixels in the image, the second histogram being divided into a second predetermined number of intensity ranges; and determining that the image is a black and white image if any one of the intensity ranges in the second histogram contains a number of pixels equal to or greater than a fourth predetermined threshold value, or determining that the image is a gray image otherwise.

Claim 6

A method of determining color composition of an image, the method comprising: calculating an intensity value and a saturation value for each pixel of the image; comparing the calculated intensity and saturation values for each pixel with first and second predetermined threshold values, respectively; labeling pixels with calculated intensity values above the first predetermined threshold value and calculated saturation values above the second predetermined threshold value as color pixels; applying a mask to the image and counting the number of color pixels out of the pixels selected by the mask; determining that the image is a color image if the number of color pixels selected by the mask is greater than or equal to a predetermined density value; and after determining that the image is not a color image, the method further comprises: calculating a first histogram of the intensity values of all of the pixels in the image, the first histogram being divided into a first predetermined number of intensity ranges; choosing an intensity range in the first histogram containing the greatest number of pixels; setting a median value of the chosen intensity range as a background value for the image; updating the intensity values of the pixels in the image by performing a dilation function if the background value is greater than a third predetermined threshold value or performing an erosion function if the background value is less than or equal to the third predetermined threshold value; calculating a second histogram of the updated intensity values of all of the pixels in the image, the second histogram being divided into a second predetermined number of intensity ranges; and determining that the image is a black and white image if any one of the intensity ranges in the second histogram contains a number of pixels equal to or greater than a fourth predetermined threshold value, or determining that the image is a gray image otherwise.